ShareefBaba Shaik
Building intelligent autonomous robots where AI meets the physical world — from perception and navigation to full system deployment.
Engineer by training.
Builder by obsession.
I'm a Robotics Engineer at Tensrai, Hyderabad, building autonomous systems that operate in real industrial environments. My work sits at the edge of what robots can do today — deploying navigation stacks on 2500 kg payload AMRs, solving problems that don't have clean textbook answers, and shipping systems that actually run in warehouses.
My background in Mechanical Engineering gives me a lens most software engineers don't have — I understand the physics of the machine, not just the code that drives it. That means I can debug a localization failure at the sensor level, tune a costmap for a specific floor surface, or design a docking strategy that accounts for real mechanical tolerances.
I work across the full robotics stack — and I mean full stack. On the robot side: LiDAR perception, SLAM localization, Nav2 motion planning, YOLO and Vision-Language Models. On the infrastructure side: I architected and wrote all backend code for our robot monitoring dashboard — AWS EC2 deployment, WebSocket communication between ROS2 and the cloud, IP management, data flow optimization, and latency tuning to keep real-time telemetry reliable across the network. From the robot's sensors to the operator's screen, I own the entire pipeline.
Outside Tensrai, I'm building an open-source research pipeline on robot learning — collecting cross-embodiment manipulation datasets in simulation and working toward a publishable contribution. I believe the most interesting problems in robotics today are at the intersection of AI and physical systems, and that's exactly where I want to be.
Tech DNA
Full-stack robotics
What I do best
Building autonomous navigation stacks for heavy-payload AMRs — custom Nav2 plugins, precision docking, and real-world deployment.
Vision and sensors
Mapping the world
Languages and tools
Engineering roots
Virtual testing
Ship to production
Featured work
Cross-Embodiment Manipulation Research — Robot Zoo
Built an open-source multi-embodiment manipulation dataset in NVIDIA Isaac Lab — 13,058 demonstrations across 3 fundamentally different robots (Franka Panda, Stretch 3, Fetch) performing 5 manipulation tasks. Designed a BaseRobot abstraction so each robot shares a common interface despite completely different control strategies. Dataset live on HuggingFace in LeRobot format, ready for imitation learning research.
Autonomous Local AI Agent — Zero Cloud, Full Stack
Built a fully private, autonomous AI agent running 100% locally on a single machine. 13 integrated layers: 5 local LLMs via Ollama, voice interface (Whisper + Kokoro TTS), 8-layer persistent memory system with ChromaDB, MCP tool servers for filesystem/GitHub/browser/calendar, computer vision and screen control, health monitoring, and a free cloud fallback router. Total monthly cost: $0.
Autonomous Mobile Robot System — Tensrai
End-to-end autonomous navigation and monitoring system for a heavy-payload industrial AMR. Built the complete robotics stack — from a custom Nav2 planning plugin and LiDAR-based perception to a full cloud monitoring dashboard with AWS EC2 and WebSocket ROS2 bridge. Currently deploying at real warehouse locations.
Mapless Navigation with ROS2 + LiDAR Odometry
Explored autonomous robot navigation without pre-built maps using NVIDIA Isaac Sim. Integrated two LiDAR sensors — Ouster 3D LiDAR (FAST-LIO/LIO-SAM odometry) and NVIDIA Rotary 2D LiDAR (Nav2 obstacle avoidance) — into a complete ROS2 pipeline. Built a working mapless system where the robot reaches goals using only real-time LiDAR-based local planning.
Pothole Detection & Repair Robot — SRM University AP
Collaborative robotics research project at SRM University Amaravathi. My contributions spanned technical design and project management — designed the robotic arm system in SolidWorks, wrote the fabrication plan, built MoveIt2 simulation and ROS2 integration, handled component procurement and fund management, and contributed to research documentation.
Mobile Robot with 2-DOF Manipulator Arm
Simulated a mobile robot with a 2-DOF manipulator arm in Gazebo and ROS2. Built the URDF/XACRO models from scratch, set up TF frames, integrated a depth camera, and visualized everything in RViz2. A hands-on exercise in combining mobility and manipulation in simulation.
Electrical Power Generation Using Speed Breakers
B.Tech final year capstone project. Conceptualized, designed, and implemented a system to harness kinetic energy from vehicles passing over speed breakers and convert it into electrical energy to power street lighting — contributing to energy-efficient and sustainable urban infrastructure.
Let's talk
Whether you're building a robot, looking for a collaborator, or just want to geek out about autonomous systems — I'm always up for a conversation.